Q & A > Question Details
In our refinery in India after revamping of Hydrocraker unit for increase of Hydrocraker unit load to by 30% than earlier, Fractionator column perfomance seems not stabilize at all. There was minor modification done in fractionator column for increase of 30% load by replacing some earlier tray with high capacity valve trays & providing some packing ring inside column. But we are now facing problem of high column pressure of around 1.45 Kg/cm2 against design of 1.1 Kg/cm2. For that column remains upset most of the time as naphtha is not removed properly. Our product flash point is also found lower than design. To compensate for flash point & removal of naphtha from product we always kept column top temeprature at higher side of 99 0c than design 93 0c. For that our high naphtha production always remains a concern as light end section found upset. It also observed that our column feed inlet temeprature always kept slightly lower @360 0C than design of 374 0c. Presently we bring down the column pressure to nearly design pressure of 1.1 kg/cm2 by running two offgas compressor. Now I just want to know can we now keep column top temeprature 93 0c as per design for less naptha production and also meet the flash point requirment of products by increasing the column feed inlet to design 374 0c.
 
Answers
06/08/2012 A: Rajkumar Chate, Sulzer, rajkumar.chate@sulzer.com
Increasing the feed temperature from present 360 deg C to design value of 374 deg C is not going to help you in higher naphtha production, boiling range of naphta is still 220 deg C at 760 mmHg if your feed temperature is 20% excess than the boiling range of naphta then at flash zone naphtha is already in vapor state and it is not going with the bottom stream. If you are loosing the naphtha in bottom stream then increasing the feed temperature will help to pull the naphta. Insted of increasing the feed temperatue play with PA rates and return temperature and top reflux.
In this case i will recommend to compare the present performance with the earlier one (before replacing the column internals, you might be having a hydraulics data sheet chek jet flood, DC velocity, DC backup and dry drop values. If your hydrualics data sheet show all the hydraulics parameters ok then there is less possibility that this problem is due to change in internals. One thing i also like to tell you that please chek in which section you have replaced trays with packing, if your naphtha fractionation section is replaced by packings then definetly it will create a problem. packings are not good for light component separation, due to low density difference of liquid and vapor phase for lighter component packing will give worse separation, this can also be the reason that you are not meeting the flash point of naphtha.
Better to collect data then simulate the tower and once agian check the hydraulics with generated vapor liquid loadings. Earlier your column was operating fine but just after replacing the trays u started facing this problem means there is something wrong with internals, to check this you can reduce the throughput and chek the pressure drop and naphtha quality, if with reduced throughput you get better results then its a problem in internals.
13/07/2012 A: Satyalal Chakravorty, Sr Consultant, satya1354@yahoo.co.in
It is a case of high liquid loading resulting liquid back up inside the column. All actions related to such phenomena to be analysed to come to a conclusion, which includes estimation of down-comer loading, possibility of blockage inside trays if the slot velocity in valve trays are in acceptable range.
13/07/2012 A: Ralph Ragsdale, Ragsdale Refining Courses, ralph.ragsdale@att.net
First, a simulation is performed to determine the tray loadings and other important parameters. But in general, increasing heat input anywhere will increase internal reflux in the naphtha section and improve the fractionation, that is, unless the internals are overloaded or out of place. I had one revamp experience where a new section was added to the top of a column. During precomissioning checkout, it was noticed that the bottom of the downcomer from the lowest new tray touched the tray deck of the old top tray in the section below. If we had not corrected the problem before startup, well, one can envision no downward flow in the column at all.